Juggling Holy Grails on His Desktop

  • Share
  • Read Later
I've been looking for aliens for at least eighteen months straight.

I'm talking about the SETI@home project, that enormously ingenious creation of Berkeley University's eternally optimistic Search for Extra-Terrestrial Intelligence department. Two years ago, needing extra supercomputing power to sift through all the noise the night sky throws at us but faced with horrific budget cutbacks, the eggheads turned to the then-radical concept of distributed computing.

All you had to do was download a chunk of software that would work as a screensaver, and your computer would effectively become a part-time worker for Berkeley, processing the data from your own tiny bit of the sky. You got pretty colors and a sense of doing good; Berkeley got what became the world's largest parallel processor. It succeeded beyond the professors' wildest dreams, and not just because the screensaver became the fashion item du jour among every Trekkie geek in Silicon Valley.

But here's my dilemma: in recent months, I've become increasingly aware of a rival for my computer's spare time. A cancer research project from United Devices (ud.com) run in conjunction with aforementioned laggards Intel and Oxford University, my alma mater. What it does precisely is difficult to understand if, like me, you do not hold a degree in molecular biology. But basically it uses your PC's power to try out thousands of molecule combinations to see if they will add up to the right sort of drug -- the kind that will inhibit or one day even cure the big C.

The moment you hear about it and realize how easy it is to download, it seems rather a good idea. Such a phenomenally useful idea, in fact, that for anyone to not be running this program on their computer — or running something else like an alien search — suddenly becomes a monstrous injustice. How can you look at those little orange pills on the cover of this week's TIME and not want to contribute to the creation of their successor? How can you think of the five million people killed by cancer in the last 11 years and not want to devote your PC's entire existence — heck, to buy a whole raft of PCs — and possibly save countless future existences? Go too far down this philosophical road and you'll end up like Oskar Schindler at the end of Schindler's List, clawing weepily at his overcoat because he could have cashed it in to save one more life.

Of course, for all I know running the UD cancer-cure program could come up with a billion useless molecules over the next decade, while Seti@home tomorrow could zero in on the signal that alien race that's being trying to hail us for centuries so it can send us some info on how to wipe out all known diseases. As with just about anything in modern science, the layperson is going to encounter impossible knowledge barriers in trying to seriously assess the value of one discipline over another. Right now, for example, my UD program is piecing together the 3-D model of a handsome little molecule, containing carbon, oxygen and sulfur atoms and known memorably as 1-3-17-35-7. Its current protein target is superoxide dismutase.

Meanwhile, on my other computer, Seti@home is busily examining 0 hour and 50' plus 26 degrees from Arecibo Radio Observatory on the night of January 1 2001. Its base frequency is 1.42804 GHz. I do not know what any of this means, nor am I ever likely to. I have voluntary ceded partial control of my processors in exchange for general-purpose, karma-positive eye candy. All I know is I've got colorful molecular shapes floating around on one screen and an awesome spectrum of random colors whizzing across another.

And that, I believe, that absolves me from the ethical choice of one over the other (if you've only got the one computer, try time-sharing screensavers). Here's what else I know: distributed computing is huge. It is the future of underfunded, highly important science projects. But eggheads beware: the online do-gooder market is not boundless. There is a danger that you might saturate the market, create a plethora of worthy-cause software, and lead consumers into a real ethical paralysis: do I help conduct AIDS research in my lunch hour or devote computer time to the resettlement of earthquake refugees?

Perhaps what we need is a National Distributed Computing Council to create one piece of client software for us all, which will divide its awesome unified power randomly to a new cause every day. In the meantime, I'm keeping my fingers crossed that tomorrow I'll discover alien life and a cancer cure, and maybe stumble on a perfect HIV-inhibiting molecule for the trifecta.

Then, after I'm rich and famous, I'll be able to afford a supercomputer of my own.